Synthesis gas containing hydrogen and carbon monoxide is produced for a variety of industrial applications, for example, the production of hydrogen, chemicals and synthetic fuel production. Conventionally, the synthesis gas is produced in a fired reformer in which natural gas and steam is reformed in nickel catalyst containing reformer tubes at high temperatures (e.g., 850° C. to 1000° C.) and moderate pressures (e.g., 16 to 30 bar) to produce the synthesis gas. The endothermic heating requirements for steam methane reforming reactions occurring within the reformer tubes are provided by burners firing into the furnace that are fueled by part of the natural gas. In order to increase the hydrogen content of the synthesis gas produced by the steam methane reforming (SMR) process, the synthesis gas can be subjected to water-gas shift reactions to react residual steam in the synthesis gas with the carbon monoxide.
A well established alternative to steam methane reforming is the non-catalytic partial oxidation process (POx) whereby a substoichiometric amount of oxygen is allowed to react with the natural gas feed creating steam and carbon dioxide at high temperatures. The high temperature residual methane is reformed through reactions with the high temperature steam and carbon dioxide.
An attractive alternative process for producing synthesis gas is the autothermal reformer (ATR) process which uses oxidation to produce heat with a catalyst to permit reforming to occur at lower temperatures than the POx process. Similar to the POx process, oxygen is required to partially oxidize natural gas in a burner to provide heat, high temperature carbon dioxide and steam to reform the residual methane. Normally some steam needs to be added to the natural gas to control carbon formation on the catalyst. However, both the ATR as well as POx processes require separate air separation units (ASU) to produce high-pressure oxygen, which adds complexity as well as capital and operating cost to the overall process.
When the feedstock contains significant amounts of heavy hydrocarbons, SMR and AIR processes, are typically preceded by a pre-reforming step. Pre-reforming is a catalyst based process for converting higher hydrocarbons to methane, hydrogen, carbon monoxide and carbon dioxide. The reactions involved in pre-reforming are endothermic. Most prereformers operate adiabatically, and thus the pre-reformed feedstock leaves at a much lower temperature than the feedstock entering the pre-reformer. Another process that will be discussed in this invention is the secondary reforming process, which is essentially an autothermal process that is fed the product from a steam methane reforming process. Thus, the feed to a secondary reforming process is primarily synthesis gas from steam methane reforming. Depending on the end application, some natural gas may bypass the SMR process and be directly introduced into the secondary reforming step. Also, when a SMR process is followed by a secondary reforming process, the SMR may operate at a lower temperature, e.g. 650° C. to 825° C. versus 850° C. to 1000° C.
As can be appreciated, the conventional methods of producing a synthesis gas such as have been discussed above are expensive and require complex installations. To overcome the complexity and expense of such installations it has been proposed to generate the synthesis gas within reactors that utilize an oxygen transport membrane to supply oxygen and thereby generate the heat necessary to support endothermic heating requirements of the steam methane reforming reactions. A typical oxygen transport membrane has a dense layer that, while being impervious to air or other oxygen containing gas, will transport oxygen ions when subjected to an elevated operational temperature and a difference in oxygen partial pressure across the membrane.
Examples of oxygen transport membrane based reforming systems used in the production of synthesis gas can be found in U.S. Pat. Nos. 6,048,472; 6,110,979; 6,114,400; 6,296,686; 7,261,751; 8,262,755; and 8,419,827. There is an operational problem with all of these oxygen transport membrane based systems because such oxygen transport membranes need to operate at high temperatures of around 900° C. to 1100° C. Where hydrocarbons such as methane and higher order hydrocarbons are subjected to such high temperatures within the oxygen transport membrane, excessive carbon formation occurs, especially at high pressures and low steam to carbon ratios. The carbon formation problems are particularly severe in the above-identified prior art oxygen transport membrane based systems. A different approach to using an oxygen transport membrane based reforming system in the production of synthesis gas is disclosed in U.S. Pat. No. 8,349,214 which provides a oxygen transport membrane based reforming system that uses hydrogen and carbon monoxide as part of the reactant gas feed to the oxygen transport membrane tubes and minimizes the hydrocarbon content of the feed entering the permeate side of the oxygen transport membrane tubes. Excess heat generated within the oxygen transport membrane tubes is transported mainly by radiation to the reforming tubes made of conventional materials. Use of low hydrocarbon content high hydrogen and carbon monoxide feed to the oxygen transport membrane tubes addresses many of the highlighted problems with the earlier oxygen transport membrane systems.
Other problems that arise with the prior art oxygen transport membrane based reforming systems are the cost of the oxygen transport membrane modules and the lower than desired durability, reliability and operating availability of such oxygen transport membrane based reforming systems. These problems are the primary reasons that oxygen transport membranes based reforming systems have not been successfully commercialized. Advances in oxygen transport membrane materials have addressed problems associated with oxygen flux, membrane degradation and creep life, but there is much work left to be done to achieve commercially viable oxygen transport membrane based reforming systems from a cost standpoint as well as from an operating reliability and availability standpoint.
The present invention addresses the aforementioned problems by providing an improved process for making synthesis gas using a reactively-driven oxygen transport membrane based system, which consists of two reactors that can be in the form of sets of catalyst containing tubes—reforming reactor and oxygen transport membrane reactor. Partial oxidation and some reforming occurs at the permeate (catalyst containing) side of the oxygen transport membranes and a reforming process facilitated by a reformer catalyst occurs in the reforming reactor in close proximity to the oxygen transport membrane reactor. The partial oxidation process, which is exothermic, and the reforming process, which is endothermic, both occur within the oxygen transport membrane based system and thus have a high degree of thermal integration so that heat released in the oxidation process supplies the heat absorbed by the reforming process.
Specifically, the improvements to the reactively-driven oxygen transport membrane based system include modifications to the reactively-driven oxygen transport membrane based system to carry out both a primary reforming process in a catalyst filled reforming reactor as well as a secondary reforming process within the catalyst containing oxygen transport membrane reactor.
Additional improvements to the reactively-driven oxygen transport membrane based system include modifications to the steam and hydrocarbon feed stream and downstream conditioning of the synthesis gas. In addition, using a reactively driven oxygen transport membrane reactor with hydrogen and carbon-monoxide as a portion of the feed produces a higher oxygen flux compared to reactively-driven oxygen transport membranes that use only steam-methane feed The actual difference in flux performance is a function of pressure, temperature, and reactant gas concentrations. Finally, some modifications or changes are proposed to the heat recovery train to mitigate metal dusting and carbon formation issues that adversely impact system performance, reliability and durability of the system.